As a fan of biographies, I was excited to learn about A Mind at Play: How Claude Shannon Invented the Information Age. Not only is it a timely biography, this well researched and easy to read book also captures the imagination. Because Jimmy Soni and Rob Goodman take care to situate Shannon’s contributions in their cultural context, the volume encourages the reader to explore their broader implications. Claude Shannon’s legacy is no doubt of particular interest to Leonardo readers due to the range of his work. If Shannon’s training and conception of Information Theory brings the current elevation of STEM disciplines (Science, Technology, Engineering, and Math) to mind, many of his lesser known projects clearly align with projects associated with the STE(A)M (the inclusion of Art) community, although the authors never speak of STEAM, per se. These include the playful spirit evident in his ongoing tinkering with electronic toys, his multi-faceted studies of juggling, and his unicycle experiments.

So, who was Claude Shannon? Born in 1916 in Michigan, by all accounts Shannon had an ordinary childhood. Noteworthy traits included a love of math and science, a dislike of facts, and mechanical inclinations. These proclivities led him to purse a dual degree in mathematics and engineering at the University of Michigan. After Michigan, Shannon was hired by the well-connected Vannevar Bush, then at MIT and later founder of the National Science Foundation (NSF), to help with his differential analyzer. This was a mechanical analog computer that depended on combinations of equivalent equations, using a wheel-and-disc mechanism for computation. A major problem was that the equations needed to be reconstructed for every problem, in effect annihilating the very efficiency the machine was intending to add to problem solving. The resounding question was, how could it reassemble itself on the fly? Shannon, who was conversant with both symbolic logic and electrical circuitry, produced a landmark master's thesis with an innovative solution. Titled “A Symbolic Analysis of Relay and Switching Circuits,” the young Shannon tied Boolean Logic and circuitry together, conceptualizing a path where 1’s and 0’s could represent logical operators of Boole’s (AND, OR, NOT) system, with an on switch standing for “true” and an off switch for “false.”

After a brief stint at the Institute for Advanced Study (Princeton, New Jersey) Shannon joined Bell Labs to work on World War II projects. Here he found an environment that fostered cutting-edge discovery and even met a visiting Alan Turing, another key figure of the Information Age. The sections discussing the shared interests of Shannon and Turing are among the book’s high points, particularly in light of the role of computers in contemporary life. Both probed machine intelligence, feedback and programming commands, and cryptology. The authors tell us that, according to Shannon, much was also left unsaid between them. He did discuss his notions about Information Theory with Turing, but they needed to avoid cryptography because of security concerns.

Shannon published his path-breaking two-part article, "A Mathematical Theory of Communication," “the Magna Carta of the Information Age,” in 1948, in the Bell Labs journal, at the age of 32 [1]. In it he showed that no matter the source, the sender, the recipient, or the meaning, information could be efficiently represented by a sequence of bits — information’s fundamental unit and a term Shannon introduced in the paper as an abbreviation for ‘binary digits’. (One bit is the amount of information that results from a choice between two options.)  Key here is that Shannon did not devise a theory about the meaning of communications but about the optimal means of quantifying the transmission of information and a new approach to the problem of noisy channels. In other words, Shannon’s theory is not about what we communicate but rather about the transmission of information regardless of what it contains.

Shannon began to grapple with the noise problem when the communication debate centered on the movement of electricity, and communication, per se, was seen as a war against noise. Building on 19th century experimentation in this area and the work of Bell Labs researchers (e.g., Harry Nyquist and Ralph Hartley), Shannon reconfigured the problem. His new unit of measurement, the bit, added a form for quantification that was capable of accommodating the idea that information is stochastic. “It is neither fully unpredictable nor fully determined.”

As I read I, admittedly, began to think that Shannon’s work with the communication of information was sometimes oddly paradoxical in terms of human experience. If Information Theory removes the semantic component of information, cryptology turns in a quite different direction, grappling with how to conceal meaningful information within a transmission. Despite their opposite goals, they are closely connected. It is intriguing to think that Shannon was able to bridge these two areas through recognizing their similarities. In other words, he saw that language is a symbol system or a code, enabling him to work with information as an engineering problem. If Information Theory offers a universally application model that removes meaning from the problem space, cryptology, by contrast, works with communication codes invested with meaning. Noise comes into play for both but because signals intelligence is as much about code making as it is about code breaking — it is invested in separating the meaning from the “noise” for some and not others.

At Bell Labs he also worked on a number of projects that anticipate current work in artificial intelligence. His maze-solving mouse was a wonderful example. It has a mechanical brain programmed to both solve the problems the maze posed by trail and error and then to remember the solutions once solved. A short Bell Labs film of Shannon with the mouse [2] is worth viewing, particularly in light of how the Lab used it to craft an infomercial. Shannon also worked with computerized chess programming, still a mainstay in artificial intelligence investigation and grist for the debates about the degree to which computational games and computers can fully simulate our humanness.

In 1956 Shannon left his position at Bell Labs for a professorship at MIT. He liked teaching, although he remained an outlier there much as he had been within the Bell Labs culture. Many of his projects of this period, particularly the whimsical ones, are along the lines of his tinkering work while at Bell Labs. These kinds of products have correlates within the ArtScience genre. Indeed, his home, Entropy House, served as in-house laboratory, his office, and contained an all-purpose “toy room” to store and display his gadgets.

The most refreshing element of the book is that the authors do not rely on boilerplate tropes. The Shannon we meet appears fresh, rather than packaged. The explanations are clear and because the text does not talk down to the reader it is easy to develop a dialogue with it. One of the most intriguing facets of the work is how the authors puzzled out Shannon himself. Despite their conclusion that he played throughout his life, I in turn was puzzled at times by the authors tendency to elevate engineering and applications creativity, even as they clearly liked the playful nature of his whimsical creations. This wasn’t a flaw so much as a reminder that I have spent my life searching for conjunctions between art and science, and this perhaps gives me a different sense of the terrain. Indeed, it initially seemed that the authors were elevating science over art due to use of phrases like “turning art into science would be the hallmark of Shannon’s career” (p. 41) and “[i]n banishing art and ambiguity, in finding the ways in which human artifacts merely stood for mathematics” (p. 46). As I read, it became more apparent that they were asking if we can translate qualities of Shannon’s playfulness and genius — a word they use quite a bit that I tend to dislike — into our lives generally. All in all, Shannon offered them a means to pose some good questions and to think about what makes creative thinkers more than technicians.

At the end the authors sum up Shannon as a man who “tackled some of the most significant scientific questions of his era and worked at the boundaries of math, computer science, and engineering” (p. 277). Then they ask if the push for STEM studies allows for the kind of innovative, creative thinking that defined his life. They also note that, “Shannon’s admirers are just as quick to compare him to M.C. Escher or Lewis Carroll as they are to put him in the company of Albert Einstein or Isaac Newton. He turned arid and technical sciences into vast and captivating puzzles, the solving of which was play of the adult kind” (pp. 278-279).

There is a lot to unpack here in terms of STEM, the comparisons they make, and problem solving. First, I liked the comparison with Lewis Carroll, who was a mathematician professionally and had a fascination for math puzzles throughout his life [3]. Shannon’s first publication as a young student mirrored this fascination. It was a solution to a math puzzle that was published in The American Mathematical Monthly. Since Shannon was “a born tinkerer” and gravitated to making things throughout his life, I admittedly see him more in terms of the inventive, cross-disciplinary and hands-on Leonardo da Vinci than someone like M.C. Escher who, while a creative thinker, worked within a more limited framework. Not only was Leonardo an engineer and a problem solver, like Shannon he wrote code of a sort, for it is said that his mirror-writing was adopted to hide his scientific ideas from the Church where such ideas were blasphemy. Of course, Shannon worshipped Thomas Edison, a distant relative, and his legacy also comes to mind here.

Second, and by extension, I am somewhat uneasy with the idea that Shannon’s contributions are considered science since I see the thrust of his life more in terms of engineering and applications, another reason he seems more in the Thomas Edison and Leonardo mold. Perhaps the dominance of STEM these days has superseded debates about whether engineering is a science, and its correlate, is computer science a science? These debates, to my mind, are comparable to those asking if design is art, and so forth.

This brings me a third point, what is STEM and does Claude Shannon’s life offers a good entry point for analyzing its strengths and limitations? I found the book’s closing comments about STEM and creativity particularly evocative in light of how disappointing I’ve found the push for STEAM. While some STEAM advocates see adding a playful element to STEM as a worthwhile goal, the STEM to STEAM push now seems more like a slogan that conflates innumerable ideas in the abstract or perhaps a strategy to normalize a particular brand of art making within the academy. While programs that encourage cross-disciplinary approaches are a good thing, and while I support finding funding streams for projects in which artists/designers and scientists/engineers collaborate, I’m not convinced that creating institutional silos that foster these projects will in turn serve as breeding grounds for creativity. They may only normalize a particular approach to cross-disciplinary problem solving. Whether normalization would aid STEM areas or even foster creativity is too complex a topic for a short review.

Shannon’s life on its own terms shows that the relationships between the technical (normalized) and playful (creative) approaches are hard to pin down. Indeed, A Mind at Play brings to mind that throughout history we find debates among scientists, artists, and humanists asking how and why excellent technicians differ from insightful, creative thinkers. STEM’s entry into our thinking about this is an iteration of these kinds of debates. Rita Colwell, then the Director of the NSF, proposed the approach in the 1980s in response to how engineering and technology had changed our world. Then and now the acronym is typically used in terms of educational policy and curriculum development. The prevailing idea within it is that a solid foundation in science, technology, math, and engineering is good for students and good for the community, for we can build a better more prosperous future when we give students the knowledge and skills to succeed in our highly complex global world. The question that inevitably comes up is how to reconcile the tension between skills that require some measure of rote learning and technical competency with the need to cultivate creative thinking.

As an historian, it is perhaps not surprising that my mind turned to an even earlier iteration of this debate as I read. Eric Havelock’s classic Preface to Plato [4] is the work that seemed most apropos as I searched for an analog to help me conceptualize Shannon’s contributions. Havelock’s research stemmed from an interest in exploring why Plato’s philosophy largely rejected poetry. As he examined the cultural shift from the dominant, oral tradition that preceded Plato to one that elevated written communication, Havelock saw the move from an oral to a literate form as a technological shift. This earlier technological communication shift, like Information Theory, altered human communication behaviorally. Havelock explains that within a nonliterate, oral culture, stability requires the transmission of experience through memorization or mnemonic learning. Thus one function of early oral poetry was to use patterns and repetition to educate the citizens by transmitting moral and technical information in easily remembered forms. This form of coding was one of Plato’s concerns because its repetitive patterning encouraged a form of rote learning and an almost hypnotic response as listeners came to emotionally identify with the repetitive content and images within the narrative. Plato thus urged a reasonable and rational approach to educating the populace so that people did not succumb to emotional inputs. The code or symbols of a written technology works well with rational and logical ideas, but it has some difficulty in accommodating other ways of knowing, for example emotional intelligence. Or, to oversimplify, several threads within A Mind at Play brought to mind that the logic of engineering and mathematical solutions fall into the kind of paradigm that is built upon a logical foundation. Shannon’s conceptualization of Information Theory does, nor is it intended to, accommodate the complexity of what humans communicate. It may have bred our digital age but as challenges like Fake News and encryption breakdowns remind us, there are many ways to use our tools.

Unfortunately, a short review cannot detail all of the amazing details the book includes. Readers will have to find these on their own. One, for example, is that Shannon landed at Bell Labs a few times before he was finally hired full-time as a research mathematician there. Indeed, the lab’s records omit the fact that Shannon was a summer intern there in 1937. Another detail that interested me was that he did a PhD in genetics, largely prodded toward this topic by Vannevar Bush. As it turned out, his passion was not in that area. Suffice it to say that in the end, the richness of Shannon’s accomplishments show how difficult it is to come up with any one-size-fits-all characterization of our humanness. Somewhat of a loner, Shannon’s proclivities nonetheless anticipated the kinds of collaborative projects often found in our world of today. Yet, he “was the sort of person for whom the concept of ‘networking’ was distasteful when applied to anything other than telephone lines” (p. 107). He is nonetheless a figure who deserves more attention, particularly from individuals who favor a broadening transdisciplinary approach.

Finally, what can we learn from Claude Shannon? According to the authors: acknowledging his creative body of work and how it defies characterization offers a useful corrective to the urge to applaud specialization in our time. I agree with this. I’m less enthusiastic about Shannon’s view of the human mind. In an interview with John Horgan he said: “I’m a machine and you’re a machine, and we both think, don’t we?” (p. 199). Clearly his interest in artificial intelligence was evident in many of his pursuits and in the machines he built. Yet, sadly, Shannon’s life reminds us that while our equations may soar to godly levels and we can craft objects that contain a machine-like precision, our human biological components and consciousness are not reducible to equations. Although Shannon believed that artificial brains would in time surpass organic ones, the book ends by reminding us that his aspiration to surpass the biological is not a part of his personal legacy. Tragically, and perhaps ironically, Shannon developed Alzheimer's disease. There were indications in the 1980s, and he entered a nursing home in 1993. Even as his body degraded, he continued tinkering. Ultimately he lost his personal communications bandwidth entirely when he died in 2001. His dementia meant that he was not able to see the digital revolution apace during his waning years, characterized by the launch of the Internet and other communication tools far beyond what twentieth century minds conceptualized. Even still, as books like The Mind at Play remind us, Shannon’s signal lives on in what he added to our communal communications repository. He was instrumental in creating the technologies of our digital age even as his inability to fathom this demonstrates that life itself is not an engineering problem, or at least not yet …. As the authors write:

“In 1948, Shannon’s theoretical work posed as many questions as it answered. But the value of that challenge shouldn’t be underestimated. … The striking feature of his paper is the reverberation, the way in which it inaugurated an entire field of study, a body of dialogue and deliberation that would long outlive its author.… Few papers can claim an impact so enduring (it has more than 91,000 citations and counting!), and its no exaggeration to say that, though information theory has important antecedents prior to Shannon, the formal study of information begins in earnest with his work. (p. 274)”

References

1. The two-part paper is available online, see C. E. Shannon, "A mathematical theory of communication," in The Bell System Technical Journal, vol. 27, no. 3, pp. 379-423, July 1948. doi: 10.1002/j.1538-7305.1948.tb01338.x (URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6773024&isnumber=6773023) and  C. E. Shannon, "A mathematical theory of communication," in The Bell System Technical Journal, vol. 27, no. 4, pp. 623-656, Oct. 1948, doi: 10.1002/j.1538-7305.1948.tb00917.x (URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6773067&isnumber=6773065).
2. The short film is available at https://www.youtube.com/watch?v=vPKkXibQXGA.
3. See my recent Leonardo review of the Lewis Carroll Society of North America meeting, https://www.leonardo.info/review/2017/05/review-of-lewis-carroll-society-of-north-america-spring-meeting.
4. Havelock, Eric. 1982. Preface to Plato: Belknap Press of Harvard University Press.